Demystifying Salaries: A Data Science Approach to Predicting Salary Ranges

This presentation dives into the world of data science and explores its application in predicting salary ranges. We'll uncover the secrets hidden within data sets, unveil the power of machine learning algorithms, and shed light on factors that influence salaries in today's job market.
Visit for more https://bostoninstituteofanalytics.org/data-science-and-artificial-intelligence/

1. The "Salary Prediction Dataset" is a synthetic dataset generated for the purpose of exploring salary prediction tasks. It contains simulated data reflecting various factors influencing salary levels such as education, experience, location, job title, age, and gender. This dataset can be utilized for predictive modeling tasks to estimate salaries based on these factors.

2. Data Information

3. Data Insights

4. EDA Title distribution Withing Manager, Director, Analyst, Engineer. Which shows Count of Manager and Director are high followed by Analyst and Engineer “Gender Relationship” that shows the distribution of people by gender. The pie chart is divided into two slices, colored blue and orange. The blue slice, labeled "Male", represents 51.60% of the population and the orange slice, labeled "Female", represents 48.40% of the population.

5. The salary distribution for all four job titles is positively skewed. This means that there are more people towards the lower end of the salary range. Managers have the lowest median salary among the four job titles. Directors have a higher median salary than managers. Analysts have a median salary that is lower than directors but higher than managers. Engineers have the highest median salary among the four job titles..

6. Education vs. Salary: There is a slight tendency for people with higher education levels to have higher salaries. Experience vs. Salary: there is a clearer tendency for people with more experience to have higher salaries. Age vs. Salary: there is a slight tendency for people with higher age to have higher salaries. Location vs. Salary: For some locations, there is a weak positive correlation, while for others there is no correlation.

7. Label Encoding

8. Dropping Unnecessary Columns

9. Splitting Data into Dependent and Independent Variable

10. Applying Linear Regression Algorithm Dividing data into Train and Test

11. Applying Random Forest Regressor Algorithm & Decision Tree Regressor

12. Applying Random Forest Regressor Algorithm

13. Since the score given by Random forest regressor is on higher side. We will go with Random forest regressor